I’ve been a Research Scientist at Facebook Core Data Science since 2013. I’m broadly interested in mechanism design, game theory, optimization, and machine learning.
My recent work has focused on answering three questions:
  1.  How can we design auctions that are effective in practice? Our ad auction system is continually evolving as our product changes and as we better understand the preferences of advertisers and users. How can we allocate and price ads in a way that approximately achieves desirable properties (e.g., incentive compatibility, maximizing social welfare) while easily adapting to future product changes? And what features are important to these real-world problems but aren’t often addressed by theoreticians?
  2. How can we handle mechanism failures? When things go wrong—that is, when there is an error in the mechanism’s implementation, or when the mechanism’s rules have unforeseen consequences—how can we effectively detect, identify, and fix these problems?
  3. How can we predict counterfactual outcomes in real-world markets? Before we change the mechanism, we want to first know the effects of changing it, but estimating those effects is challenging because of market dynamics—an A/B test, for instance, may fail to capture participant responses that occur only over longer horizons or after the change is fully rolled out. How can we predict counterfactuals in such situations? Are there approaches that work well for particular types of mechanism changes?
Prior to Facebook, I received my PhD in Computer Science from Brown University, where I wrote my dissertation on prediction and optimization abstractions in complex markets. Before that, I received my BS in Computer Science from the University of Minnesota, where I studied autonomous trading agents and sequential decision making.
If you are interested in an internship, I’d love to hear from you—send me an email with the title “Facebook Internship” and your CV. I’ve been fortunate to have worked with a number of fantastic interns in the past:
Riccardo Colini Baldeschi (postdoctoral researcher at LUISS)
Christian Kroer (PhD candidate in Computer Science from CMU; winner of 2016 Facebook Fellowship in Economics and Computation)
Mark Hou (PhD in Economics from MIT, now a Data Scientist at Lyft)
Nisarg Shah (PhD in Computer Science from CMU; winner of the 2014 Facebook Fellowship in Economics and Computation; now an Assistant Professor at the University of Toronto)


Mechanism design, game theory, optimization, machine learning, sequential decision making.